Reading Guide to: Schutz  A  (1972)  The Phenomenology of the Social World,  Heinemann Educational Books: London

 NB I have reproduced male pronouns and possessives throughout, as in the original. Apologies for any offence

Chapter 3

Intersubjective Understanding 

No two individuals can have the same stocks of knowledge, and thus I can never comprehend the intended meanings of the other [conceived of as a full 'Thou'] in the same way as he does. However, I can assign meanings to his acts. This is done in a different way to normal perceptions of other phenomena, since some transcending is involved  (this involves recognising the other as another consciousness). For example, the bodily movements of the other are interpreted as having reference to his consciousness. This involves a certain intentional act, ascribing simultaneity  (we can observe our own acts only as an afterthought, on the basis of reflection, but it is possible to directly observe the acts of others, and merge them into streams of consciousness for a passage of time). 

Comprehending the acts of the other means linking the outward indications into a context of meaning, valid in so far as it corresponds to a context of meaning within which the other is constructing his present lived experience -- this is Weber's 'direct observational understanding'. Actually, I arrange both my and your acts in my meaning contexts, attending to experience according to my relevancies -- that is your structure of consciousness is given to me only in segments and in 'interpreted perspectives'. It remains impossible ever to get at intended meanings: I can make ‘mistakes’, for example, in assuming that your acts are thematic, whereas this may not be so for you. 

Understanding correlates with meaning -- that is understanding involves search for meaning, and we have only understood things when we've made them meaningful. In the 'natural attitude', we gain understanding of the world by interpreting our lived experiences of it. We attempt to understand the other  [that is the conscious other, the 'Thou'] in several ways:

Signs and sign systems

An indicator is linked to that which is indicated by a complex of meaning. The connection is solely in the mind of the interpreter or -- that is there is no necessary logical connection. The process involves relations of representation. Interpretation of a sign may arise through systems adequate to it whenever it represents, or as an object itself, and we can switch between these two approaches according to some metascheme itself a result of previous experiences  (119).  Signs also stand for the subjective experiences of the user, that is they have expressive functions. A full understanding therefore involves interpreting a sign according to the subject's experiences, or referring to what it is representing, or locating it in a sign system. One indication of understanding may be our ability to use the expressive function of the sign in a new context.

Meaning is attached to signs if their significance is understood by both actors in a conversation. Locating signs in sign systems can be done objectively. This involves co-ordinating to what the sign represents independently of whoever is using it  (for example mathematical symbols always mean definite operations). This involves the idealisation of repeatability. But we can also focus on the links between the sign and the experience of the actor. Here, an aura surrounds the objective meaning, which can be specified fully only when all the subjective meanings and occasions for their use are given  (124).  The full meaning of a sign, for example a word, is possible only where we have a knowledge of the mental structure of all the people using it, and something like a 'logic of everyday life'  to let us predict when different types are using different meanings. Subjective meanings are always 'occasional'-- that is affected by context, just as the meaning of an individual word depends on a whole prior sequence and syntax. This context emerges during discourse. Discourses are a project like any other act, with the motive of unifying different parts. The project may be a word, a book, or whole series of volumes, and the meaning of the project is never apparent until the project is finished. 

Conversations are conducted with regard to the other's subjective, that is interpretive and expressive, schemes.  The listener tries to guess what the speaker had in mind, and the speaker is trying to guess the impact of words and the likely success of his project. There are different levels of understanding, as above. The listener tries to establish both objective and subjective meanings, and then he may seek the speaker's 'in - order - to'  motive, to get it his project, and to fantasise a full meaning for it. Projects can also be motivated by 'because'  motives, which need not be subjectively experienced. Interpretation relies on the knowledge of the other, which in turn depends on the degree of intimacy established with the other.

For Schutz, the objective meaning of events arise from an ordering of the interpreter's experience, and thus can never be free from interpretive bias. Subjective meaning arises from the ordering of the producer's experiences. Objective meaning reflects a level of abstraction -- universal meanings divorced from flows of experience, containing idealities. Both subjective and objective meanings are therefore two ends of a continuum of meaning. The distinction between them is important in cultural sciences particularly. In other subjects, such as economics, academics are content to operate with an ideal actor acting in accordance with laws, and subjective meanings arise only when asking why particular managers or whatever did not act in this way. The generalised search for subjective meanings is not suitable for a science, and will result only in some argument for God, or some kind of animism.

Chapter 4

The social sciences use different categories of meaning than does the person in the natural attitude -- here, we take the world for granted, and experience others immediately. As soon as one questions the taken for granted, social science begins. Given that social science can be understood as a disruption of the taken for granted, it indicates in the simplest sense that social science must be based on prescientific categories.  

The social world can be divided according to the degree of direct experience of it: the most immediate level concerns our fellow beings; then the world of contemporaries  (where we can make inferences about subjective experiences, or actually interact with others); then the world of predecessors  (we cannot interact with these people yet we can get inferences about them); then our successors, who are inferrable only via idealisations. Understanding therefore consists of different types of processes according to this degree of experience.

Social action involves 'other orientation', that is if we are assuming that the other is a Thou -- denying this means we can treat the other as an object. Action can be 'other affecting', where my antecedent projects are motivated  (in order to)  bring about a conscious experience in the other through an attention to his flow of consciousness. There are different kinds of 'affecting' -- for example teaching may involve simply making sure that my ideas are present in the consciousness of the other, or my ideas may be intended to influence the other's behaviour more positively. My project involves selections from among the past actions of the other. I pursue a continuum of actions according to their results on the other. Some actions need not involve effects on the other at all, however, although all signs involve some tacit aim to affect, although this may not be intentional. The other need not reciprocate. My attempts to affect the other will vary according to the degree of anonymity of the other. 

Any observer watching us would seek the outward indications of a relationship between the other and me, the originator. For example, there might be relations between the actions of the actors which will be interpreted by imaginative reconstructions of projects for both of us. The observer will validate his understanding according to his experience of the social world and his knowledge of the character of the observed persons, both actual or typical. This means an other orientation for the observer as well. The observer will refer to accounts of correspondence of indications invoking events in his past -- he will make habitual interpretations from experience. However, the degrees of interpretability open to him will depend on his knowledge of the people being observed.  

It will obviously be easier to identify social interactions where people are simply oriented to their others than it will be to estimate effects on the other -- the only way to increased the probability of a valid interpretation is by asking actors if they are really affected.  It might also be possible to live in a face-to-face relationship, one of simultaneity, and then to reflect upon it, examining one's own past acts acting as an observer, and checking the validity of those observations against knowledge of one's own subjective intentions. 

There may be a minimal reciprocation, were the other is merely paying attention to me rather than actively seeking my meaning. This may lead to me aiming to make the other equally other- oriented, leading to an intersubjective motivational context. Here, the in - order - to motives in the speaker become because motives in the responder. This context of motivation which governs the reply is often taken for granted, as when we assume that a question immediately brings a reply. This only happens when the other is genuinely other -oriented himself, however. As we might expect, our knowledge of the respondent's context of motivations can be specific or typical according to whether he is a consociate or a contemporary or whatever. 

Various relationships with the other are possible. With a face-to-face relationship, characterised by immediacy, simultaneity of structures of consciousness and so on, any reflection on experiences means an end to the relationship  (because we can only attend to experiences once they have elapsed). Face-to-face is a ‘content-filled we-relationship’, seen as specific, unique, and in a concrete context. It may be remote or intimate, and vary according to intensity. I have both general and specific knowledge of my partner in such a relationship, and of his interpretive schemes, and this increases with exposure. Both his acts and his characteristics become multiple, and interaction tends to yield reciprocity. The environment can be guaranteed as a common one, shared in experience, which is useful in getting the interpretive schemes of the others -- we can make unambiguous references to objects within our mutual reach, and check our guesses by questions. The flexibility of face-to-face enables the disclosure of the other’s motivational contexts -- he continually adjusts to me and I participate to test his reactions, to see how his assumed motives produce what he says. I can continually compare my expectations of his behaviour with his actual behaviour, and can witness him actually deciding things. 

Direct social observation can occur where the other is unaware of me. His body still gives off expressive movements, and I can guess at his subjective meanings because he is acting now in a structure, that is I can see processes at work. However, no testing is possible, since we can make no reference to objects, and cannot ask questions to check observations. Observers cannot modify the behaviour of the other, and in extremes cannot even be sure that the movement is a planned action rather than some simple purposeless behaviour. However, three indirect approaches are available to assist in interpretation:

  1. We can remember our own similar actions, and try to use them to draw up the principles of motivation contexts, assuming that these principles are the same for the other as well. This can be done directly or post hoc.

  2. We can use our own knowledge to build up a picture of the 'customary behaviour'  of the individual, specifically or typically, and thus arrive at typical motives.

  3. We can try to infer the motive from the act itself  (asking, for example, whether a particular motive would be furthered by this Act). This involves an assumption that the effects we observe are as intended.

Generally, the validity of interpretation depends on the proximity to a face-to-face relationship. This is the only way to check on those assumptions of customary behaviour or intention. Face-to-face interaction typically does not reveal these very clearly either, but they are easier to infer because the data is more vivid. The problems are similarly present for any observers of social relationships, since the observers' interpretive schemes can be different from those of either speaker -- the attentional modifications are different and it becomes difficult to be aware of both equally. 

The third kind of relationship is one with contemporaries. The knowledge of contemporaries is always direct and impersonal, graspable only as general types of subjective experience. Contemporaries are not given to us without mediations. We may regard a contemporary as a former Thou, now changed in certain ways, or as a relation to a Thou. In both cases he can only be understood through a typification which is held to be invariant across the changes. Cultural objects can index our contemporaries:  they are related to the contemporary via an experienced relation to a known person or type  [eg ' my friend behaves like that']. All our experiences of contemporaries are therefore based on interpretative judgments. 

We extend our other orientation into a  ‘they orientation’, in an attempt to gain experience of social reality in general, of human beings and their operations in the abstract. Contemporaries perform actions with objective meanings -- that is meanings constituted by me and my interpretive schemes into a unity  [objectivated].  My interpretation is a synthesis, based on my experiences of many Thous. It is a personal ideal type, constructed from many cases, although it need not correspond with any of them. My social relations with others take place through these ideal types based on assumptions of generalisability, or in Schutz's terms 'again and again' [one kind of idealisation]. For example, one expects policemen to act in certain ways, and thus we can relate with them in an ideal way without ever knowing them as individuals. Similarly, such people are only relevant for us in so far as they correspond with our ideal type. 

Ideal types like this are also important for understanding predecessors. They are even used to interpret the Thou in face-to-face relationships, although they are more concrete and more liable to be modified by conversation. Ideal types are therefore vital to all elements, all meaning contexts all products and all actions. They are formed by lifting moments of experience out of their settings and contexts, via 'syntheses of recognition', by 'freezing'  them into an ideal type. They may be a type of person -- the personal ideal type -- or of an expressive process, an 'action - pattern type'. These are often related, as when a person is defined by their role. However, there are problems:

The first sort of problem arises when an interpreter discovers the motives for an actor by interpreting his act in an objective context. The same motive is applied to any act that achieves the same end via the same means, no matter who performs it. In this way we arrive at one typical motive for one typical act per personal ideal type. The next step is to link an agent with an act, a person who, with typical modification of attention, typically intends this typical act. This can only be done if we assume that ideal typical people are rational, that is, in certain states they logically construct actions without hesitation or doubt, that their act takes the form of a major goal, and that acts arise out of a definite passage from their experience  (that is they have because motives).  This all stems from an objective interpretation of the act: the observer imposes some unity on a person's experience around that interpretation. The personal ideal type is selected by the observer on the basis of an originally objective definition. It tries to explain actions, yet it is already determined by actions. It is linked with acts definitionally rather than empirically as sociologists believe. All the ideal type can offer is an account of experience polythetically constructed [roughly – constructed over time, by a number of synthetic acts], but this is not based on some real event, but on an act already conceived  monothetically [ ‘at a glance’, in one defining moment] by social scientists  [in other words, it spells out some complexity based on an original assumption]. It is an illusion to regard the ideal type as real -- circularity can ensue, where the ideal type seems real, and testable, seen as based on real actors, and then used to explain real actions which validate it in the first place. 

People are sometimes seen as being 'free’ to choose their actions, and this is justified by departures from their ideal types -- but all that is really involved is a manipulation of their typical characteristics (such as when we get to realise more of their complexity as they become more intimate to us). Generally, the more detail that is required the less secure the ideal type and the less likely that there will be a definite type to be assigned. Typing is performed only using certain criteria, and characteristics other than these are seen as properties attached to unique individuals. Since we can relate to others in degrees of concreteness or anonymity, there is often an uncertainty over how actual others might conform to ideal types, and how much we might need to modify our ideal types. We also have the possibility of turning they-relations into several we-relations, operationalising general types into more specific ones with greater concreteness.

Finally, types may be 'characterological', that is concrete and based on direct experience, or ‘habitual’, defined solely according to function.  Some habitual types can be collectivities, such as the state. We can be anthropomorphic about them but they're absolutely anonymous and permit no inference about subjective meaning contexts -- the action of the state must be reduced to the action of its functionaries and then understood via networks of personal ideal types. Artefacts can be understood via types of producers and users, and both are perfectly anonymous. 

Relationships between contemporaries therefore are always doubtful -- will the other correspond to my type, and will I to his?  Contemporaries assume a shared interpretative scheme, a notion extended after its experience in the nuances of face-to-face. Effects of the relationship can only be evaluated after the interaction, as there is no continual feedback. Much can go wrong double - for example if I make a mistake, it alters the other’s motivations, and now I can't correct my expectations. My knowledge of the other is modified only within a narrow range, since I type him only on certain criteria. We don't share the same environment, and I can only assume his environment is understandable from principles drawn from my environment. I can only use sign systems objectively, since subjective meanings are liable to misunderstanding. The only way the other can interrogate me in a they - relation is to look up words I used in a dictionary. Written letters are an intermediate form -- they are good for filling in objective meanings, but speech is the only way to deduce subjective meanings.

 These examples show that a subjective context may be essential for objective meanings. Difficulties arise for observers if they believe their own types in use rather than the subject, or mistake the two for each other. Types can be self-fulfilling, however especially if observers never confront subjects as real persons. Interpretive sociology must construct personal ideal types for actors that are compatible with those constructed by partners  (205) -- this is what Weber meant by 'meaning adequacy'.  Good relations exist between actors where there is a compatibility between the personal ideal types of each other, of typically conscious experiences, and of the relationship.

Chapter 5

Meaning involves a special way of attending to lived experiences, a special Act of attention, a turning of attention to an already lapsed experience, leading to a 'lifting out'  from a stream of consciousness  [this special act is indicated by capitalising it]. Meaning is both pre-predicative and pre-phenomenological  [that is existing before language and awareness of perception].  The specific meaning of an act involves a series of polythetically constructed Acts, although these are typically grasped monothetically  as a single 'meaning'.  These Acts operate on a stock of knowledge according to interests  [motives and relevance systems]. Meaning arises from a principle of unity, determined subjectively or synthesised objectively -- if the latter then it is wrong to mistake this for a subjective meaning. Motivation itself is a result of meaning contexts.

 The Problem of Scientific Knowledge in Social Sciences

The social scientist is an indirect social observer, who leaves out his own direct experiences. Husserl  (in Formal and Transcendental Logic) says that science must radically check taken for granted knowledge. Special scientific ideal types must be constructed, on the basis of scientific stocks of knowledge combined with explicit acts of judgment. These involve conscious thought rather than some pre-predicative synthesis and. The social sciences are exclusively concerned with contemporaries.  Judgments are made in accordance with scientific experience as a whole, rather than special human judgments. Methodology follows essentially formal logic with the theme of constituting objective meaning contexts out of subjective ones, by objective rating and anonymising ‘normal’ constructions or ideal types.

Scientific ideal types are therefore formed differently from those in the natural attitude, in that they refer to the body of scientific experience, and do not depend on face-to-face. However, they must possess both causal and meaning adequacy, according to Weber -- causal adequacy is required to make up for the lack of concreteness in meaning adequacy. Weber says there are three classes of meaning --  (a) that which is intended by an actor in a specific case; (b) the average meaning of groups of actors; (c) the meaning of ideal typical actors, such as Economic Man. However, these are but relations between ideal types, since science can only deal with types, albeit at different levels of specificity. Similarly, variation in verifiability between these types is important, ranging from knowledge of a friendship in a we relation to deductions of an ideal typical actor derived from observing behaviour. Real actors appear to be 'free', that is acting from spontaneous activity, but ideal types are always predictable -- science can only permit actors to be ideal types, as a result, [but then verification can only be a matter of consistent relationships between types].

Weber says a type is [meaning] adequate if it is constructed from 'typically relevant' acts of persons, that is originating from invariant motives, themselves typified, and featuring an ideality of 'and so forth’. Weber says motives offer a context of meaning apparent either to the actor or the observer, but there are important differences for Schutz:

  1. Direct social observation means the observer is assuming that acts indicate a project, which enables him to derive motives. However, the ‘span'  of a project can be known only to the actor -- it is common to ask actors , and then find that their apparently so relevant behaviour is not intended at all, for example.

  2. In indirect social observation, the meaning context of the observer and the actor is only ever the same if the observer assumes that the actor corresponds to a personal ideal type, that is he has a typical motive, for example.  Such similarity is not really surprising if the ideal type is constructed to produce it in the first place.

The concept of causal adequacy in Weber's work refers to the probability that the sequence will always actually occur in the same way, involving generalisations based on experience. A correct causal interpretation arises when the relationship between actions and motives is correctly apprehended, enabling such generalisations, but the relationship needs to be meaning fully comprehensible too -- that is, its component parts must constitute a typical complex of meaning. Without such meaning adequacy, generalisations remain incomprehensible, mere statistical probabilities. Without causal adequacy, however, there is no proof that the action normally in fact takes its stated course. Statistical associations become sociological generalisations only when the associated items index understandable subjective meaning.  Correspondingly, meaningful interpretations become sociological only when they provide observable indicators. However, in the drive to construct such indicators, sociologists  should not forget that measurability is 'not always directly proportional to the clarity of subjective interpretation'. 

Pursuing  this further, causal adequacy must involve some accordance with past experiences, but this involves no recognition of the other as fully conscious. Since this would be an unwise assumption, in the social sciences, we do not find causal necessity in the strict sense, but rather a 'causality of freedom'  [this seems to involve an explanation of patterns in social life as a result of voluntarily-acting individuals holding the same ends and possessing a limited means to achieve them]. Schutz agrees with Weber on the need for some 'postulate of coherence of experience', but prefers to see this as an interpretational scheme of rational action developed by sociologists, who are seeking typical ends and means, and therefore determining typical actors' choices of goals and projects. All sorts of typifications are involved, in fact, for example means it must be appropriate to goals in the light of their own experience. Thus causal adequacy is based on typically comprehended meaningfully adequate relations, and not on laws as such: in the social sciences, causal adequacy is only a special case of meaning adequacy  (quite unlike the natural sciences, where phenomena are  'beyond rational understanding'). 

Turning  to meaning adequacy, Weber's definition means that an action is meaningfully adequate where it is located in an objective context of meaning, that is a typical context, based on some average 'habits of thought'. The actor himself may assign quite a different meaning to his action. Sociologists acquire a knowledge of the 'average habits' through their own personal ideal types. Meaning adequacy therefore refers to problems of interpretation of concrete actions via ideal types, which must not contradict what else we know about the actor, knowledge gained through other ideal types. The whole notion of meaning adequacy is therefore irreducible to the proviso that our interpretation should not contradict previous experience. This is exactly the same as the principle underpinning causal adequacy!  This can explain the common experience that observers often ascribe causal adequacy and then have meaning adequacy revealed to them -- in reality, both are already present in the typification of the actor. It is not surprising that from the point of view of a single observer, either both types of adequacy are present or neither. Nor is it surprising that Weber tells us that ideal types must be: adequate on both levels  (that is they must not feature any atypical behaviour);  compatible with our experience of the world and of other people  (and of this person in other situations);  based on repeatable behaviour; sufficient to explain the action without contradicting previous experience. 

Weber does say that it is possible to adequately explain behaviour without regard to the subjective experiences of the actor. This is objective probability [and is found in studies such as Durkheim's Suicide]. Subjective probability is a matter for the actor, and it turns on expectations that future acts will succeed. Subjective probability is assigned to projects and 'in - order - to'  motives. 'Because motives' can only have objective probability.  Subjective probability leads to a theory of rational action if the actor is assumed to have based it on a series of judgments about his goal, and the possibilities of attaining it, that is if he is assumed to regard experiences typically in a mode of attention, rather than passively  (the latter involves action which is 'driven', for example by neutralised habits).

Actions involve projects, so they must be assumed to be rational in regard to a goal. There are some complications though: for example, means and ends are found in hierarchies, so that ends become means to some higher goal;  goals may be well-specified, while means are left vague, or vice-versa. These complications lead to problems for ideal types, since these are based on clear ends and means, fixed motives, and 'pure'  means based on adequacy, and chosen so as to give actions the maximum chance of success. Once we have assumed rational action of this kind, it is easy to convert objective meaning and probability into subjective meaning and probability, especially where universal problems are being considered  (such as those normally thought of as 'drives'). Irrational action, where there are confusions, or lack of specification, can be seen by sociologists as a deviant type, produced by adding peculiar 'in - order – to 'motives, as in Weber's 'affectual'  and 'traditional'  types in his work on religion. However, investigation of these deviant types must itself always be rational, using logical interpretive schemes, as in all social sciences  --  Schutz is completely against interpretation based on 'metaphysics and intuition'  [one of his targets is Dilthey]. 

Thus all social science features meaning contexts themselves based on subjective meaning contexts. Knowledge is always indirect, and never of some immediate social reality.  Actors are always personal ideal types, never a Thou with full consciousness (duree [roughly, a sense of duration, of subjective time] and spontaneity). Sociological Man experiences only those things which are consonant with already prescribed motives -- he is assigned experiences on the basis only of adequacy, and these must be related to the experience of the observer, and be compatible with each other. 

Interpreters have their own typifying methods, which try to replace subjective with objective meaning at various stages. These produce a 'conceptual model, not a real person'  [an 'homunculus' in another of Schutz’s phrases]. The sociologist’s types vary on the dimension of anonymity and concreteness, and they can also be used to describe objects or courses of action.  They may be constructed from many kinds of experience, including 'empirical', or 'eidetic'  (that is, from insights and imagination). They are constructed from such experiences by processes of abstraction, a generalisation, and formalisation. Even 'laws', as in economics, are based on such prior typifications, at an extremely anonymous and abstract level. The type in use defines rational actions, and deviations from the rational. If concrete persons appear to engage in atypical action, this can be dismissed as irrelevant, again seen best in economics  [such atypicality is commonly banished to some ‘unknown’, ‘ransom’ or noninterpretable action in empirical work, of course]. 

All the sciences decide what is relevant by referring to some higher interpretive schemes, which tell them which constructs may be used  [paradigms, as in Kuhn, perhaps?]. Sciences can also be classified according to whether they deal with form  (and examples here include structuralism and systems theory), or actual content  (the 'already given historical or social acts').  The particular role of interpretive sociology, however, is to look up processes of meaning establishment and meaning interpretation by individuals, at all levels.